Hacker-in-Residence: Flying Robots


By now, our Hacker-in-Residence program is humming right along, and we mostly stand around tapping our fingers together with glee while some excellent applications roll in. One such excellent application came from our new friend John Gillespie, and we’re happy to have him join us for the next two weeks!

alt text

The man, the myth, the legend: John Gillespie

John’s favorite movie is “The Adventures of Buckaroo Banzai Across the 8th Dimension,” his favorite snack is yogurt-covered raisins, and he claims he can win any carnival game you put in front of him, as long as it’s not rigged. Important statistics aside, we also asked John about his background and plans for his residency.

Tell us your background, interests, and some favorite past projects. What and where is your current position/company?

My background is in mechanical engineering and programming. I am married with three kids and live in a tiny town in the Texas hill country. I am the founder & lead developer for a small mobile development company - Product7. I have been developing for the iPhone for over four years - I have over 20 apps in the Apple app store - mainly in the education market.

A recent fun project I worked on was at the AngelHack mobile AppHack in Austin, TX, in early November. I won 2nd place with ShareMesh – an iOS based encrypted peer to peer messaging app that does not require an internet connection to operate. The app used the new iOS 7 features iBeacon and Multipeer Connectivity to handle discovery and communication between devices. I had a great time and there were a lot of interesting projects and very talented people working on them.

How and why did you get involved in SparkFun’s Hacker in Residence program? Why do you think programs like this hacker-in-residence are valuable?

I think the next big thing is the intersection and hardware and software (internet of things), and I am very interested in robotics and the DIY drone & UAV efforts, especially those based around the Arduino. I love writing code but I am definitely a newbie with regards to hardware projects. I have been using SparkFun products and tutorials to help me ramp up on electronics and the Arduino over the past few months. I saw the Hacker in Residence program and applied - the biggest benefit to me is the access to SparkFun’s expert engineers and resources, plus the time and focus to start and finish my project.

What is the project you’ll be working on at SparkFun, and how long will you be here? Why did you choose this project?

I want to create a single propeller flying UAV that is as maneuverable and stable as a quadcopter. The single larger rotor is more efficient, quieter, and should be cheaper to build as well. I’ll use software and aerodynamic forces (ailerons & rudders) to keep the single prop design stable in hover.

I chose this project because I am very interested in solving real world problems with ‘swarms’ of simple flying robots. Before doing that I need to as a first step to design extremely low cost basic flying robots that eventually will work together to perform tasks in the real world. I’d like to see the world of robotics focus on generic, modular, and cheap hardware that can be re-configured by software on the fly to handle different conditions and/or tasks - instead of expensive hardware customized for every task.

My initial goals are to develop a single rotor ‘tail sitter’ UAV that can sit in stable hover indoors at a designated altitude over the floor with no user input (software controlled), and to update code/hardware to handle external inputs (wind from a fan, push from person) and regain stability. My stretch goals are to create code/hardware so that the drone will follow the user (without obstacle avoidance) within a certain distance, probably using iOS beacon tech (Bluetooth LE).

We’re excited to have John here and can’t wait to see what he ends up building. If you’d like to apply to be a SparkFun Hacker-in-Residence, you can apply here!


Comments 12 comments

  • So, something not entirely unlike https://en.wikipedia.org/wiki/Radio-controlled_aircraft#3D_flight ?

  • Glad to see a fellow Texan as hacker in residence and even happier to see you working on personal UAV tech. Our Legislature passed a really weird law on UAV’s this year regulating them through criminal statute and it was clear they had no idea about their non-nefarious uses. Texas can and should be a leader in drone tech. We just need our techies to lead and show them what’s possible because our politicians are pretty clueless.

  • Personally, I would probably tackle this as far as electronics is concerned in two steps. I would make it RC, with the output for rudder, elevator, throttle and ailerons fed into a control module. The control module would then drive the servos given the current inputs form the sensors. That would allow the basic design for the control module to be debugged. Then once stable flight and manourvering can be done, then add a autonomous flight module that feeds the control module. Maybe in the future combine into one module.

  • not to be snarky, but aren’t you describing a helicopter? i understand the hovering and following additions you want to make, but the mechanics sound like a helicopter.

    which, of course, is a successful design and i’m all about using off the shelf stuff…

    • They both have propellers on top of them, but there are fundamental differences between a helicopter and a tail-sitter. In particular, helicopters have complex hubs that can change the pitch of the blades independently at various points in the rotation, which provides lateral control. A tail-sitter has a simpler propeller system and relies on control surfaces in the slipstream for lateral control.

      • Coaxial fixed-pitch flybarless rotors is what I would use, they are very stable and efficient, no tail rotors or anything. I have a couple of collective pitch helis, the largest is a 500-size with a rotor span of about a meter. Rotors are so much more beautiful than props.

      • Also be interesting to see what the size and shape of the body needs to be to counteract the torque of the motor. The picture that you linked too uses counter rotating props to counteract the torque. Model 3D planes can hover vertically, but they tend to use small, higher speed props ( less torque ) and are a full plane, with around a 36" wingspan. A big prop, needs a big motor and lots or torque.

        • Having enough lift from a single prop & motor to overcome the weight and enough force X displacement vector of the airfoils in the propwash to counteract the torque of the motor is the challenge. I wonder if instead of multiple flight control surfaces let the airframe have some rotation. This way all except one of the airfoils in the propwash could be fixed. At each position in the rotation the microcontroller adjusts the angle of attack of the control surface creating lateral movement by favoring one side. The airframe rotation would need to be slow enough to read the gyro and adjust the control surface but fast enough to give the maneuverability that you want.

          • What you just described is the electronic version of a swashplate !

            For the body, I was first picturing in my mind two inverted U frames at 90 degrees to each other. The bottom of the U would be the legs, and the recess in the middle allowing a payload to be carried. The legs would also be built up with decent area, to provide the necessary drag to stop the spinning of the body. This part would carry the controls.Then I realised there would be no controls in the propwash.

            Each time I thing about the design to solve the body spinning issue, I always end up with something that resembles a fixed pitch helicopter

            • Not quite a swashplate. Similar principle but I was thinking a fixed pitch main rotor and allowing the body to rotate. The body would have airfoils in an X or starburst pattern. The airfoils would have the leading edge facing the airflow of the rotor, countering the rotation and with some control providing the maneuverability.

    • Tail-sitters have conventional wings for more efficient horizontal flight. However, it doesn’t sound like transitioning to horizontal flight is in the plan.

  • John I’m pretty excited to see the results you get. I’m working on quadcopter design, but would really like to see what your single rotor concept will yield. I was just watching what Amazon wants to do with delivery drones and I think that that market will blow up soon for deliver systems, search and rescue and just about every other use that can be thought of for an arial assisted system. Good luck and can hardly wait to see your progress.


This Week

This Month

Heartbleed

Happy Arduino Day!

SXSW Create 2014